Guest a_appleby Posted July 23, 2011 Report Share Posted July 23, 2011 No, i think the gpio repeat rate is not the problem (it can be configured...) ... the problem seems to be a bug in the pwm controller driver of tegra. By commenting those lines in the shuttle-*-gpu.c, the fb console is never enabled again, so we need those lines (the calls to fb_blank()). But, enabling those lines triggers the pwm bug. I think the best approach will be to fix the tegra pwm driver to accept that multiple enabling/disabling... And i also think this bug should be fixed in later versions of the .36 kernel... will also check that. I will take a deeper look into this and the wifi problem in a few hours. Thanks for your patience! :D Regards, Eduardo I'm looking forward to these changes. ;) I need to say you are great . ;) What ROM are you using now? I couldn't get sound to work on vegatest and on the system.img you provided. Please share it. Link to comment Share on other sites More sharing options...
Guest newbe5 Posted July 23, 2011 Report Share Posted July 23, 2011 I'm looking forward to these changes. ;) What ROM are you using now? I couldn't get sound to work on vegatest and on the system.img you provided. Please share it. Hi! OK, so to keep people up to date, here is the test platform that myself, rebel1 and Corvus are currently working on: Vega Test22 In this build, Audio is working, as is the SDCard, although Vold is being it's usual pain in the ass, so you have to unplug the SDcard and plug it back in post-boot for it to be picked up. This build is messy as hell, but it's more about testing the .36 changes than making an actual functional ROM right now. As always, if you fix anything, please share! ;) newbe5 Link to comment Share on other sites More sharing options...
Guest a_appleby Posted July 23, 2011 Report Share Posted July 23, 2011 Hi! OK, so to keep people up to date, here is the test platform that myself, rebel1 and Corvus are currently working on: Vega Test22 In this build, Audio is working, as is the SDCard, although Vold is being it's usual pain in the ass, so you have to unplug the SDcard and plug it back in post-boot for it to be picked up. This build is messy as hell, but it's more about testing the .36 changes than making an actual functional ROM right now. As always, if you fix anything, please share! ;) newbe5 Thanks! Will report asap. Does video acceleration work on this one? I am talking about AVP related errors in the dmesg. Link to comment Share on other sites More sharing options...
Guest BlackTigerX Posted July 23, 2011 Report Share Posted July 23, 2011 (edited) The whole process of getting the .36 kernel to work has gone far beyond my capabilities :rolleyes: The only thing I could help with now is testing your builds - I have the Smartbook Surfer 360 MN10U which is the Advent Vega with integrated 3G+GPS. I really like this progress in here. Awesome job you guys do here! Edit: I just tested newbe5 late'st test build and although it's supposed to be messy, it's in fact is very fast! Is this speed increase (compared to VecaComb 1.7) related to the new kernel - or is it simply because only the most important apps are included!? Edited July 23, 2011 by BlackTigerX Link to comment Share on other sites More sharing options...
Guest a_appleby Posted July 23, 2011 Report Share Posted July 23, 2011 Hi! OK, so to keep people up to date, here is the test platform that myself, rebel1 and Corvus are currently working on: Vega Test22 In this build, Audio is working, as is the SDCard, although Vold is being it's usual pain in the ass, so you have to unplug the SDcard and plug it back in post-boot for it to be picked up. This build is messy as hell, but it's more about testing the .36 changes than making an actual functional ROM right now. As always, if you fix anything, please share! ;) newbe5 Since Eduardo is working on the wifi / wake from sleep, I tried to add Camera.apk back and debug the camera, but it wouldn't show up in the list of apps no matter where it was stored. Terminal emulator was FC-ing from /system/app, but it worked from /data/app. What should I copy from this build to the system.img from rebel1 to make video / 3D acceleration work there? I am talking about the AVP errors. The whole process of getting the .36 kernel to work has gone far beyond my capabilities :rolleyes: The only thing I could help with now is testing your builds - I have the Smartbook Surfer 360 MN10U which is the Advent Vega with integrated 3G+GPS. I really like this progress in here. Awesome job you guys do here! Edit: I just tested newbe5 late'st test build and although it's supposed to be messy, it's in fact is very fast! Is this speed increase (compared to VecaComb 1.7) related to the new kernel - or is it simply because only the most important apps are included!? It's the kernel and the video driver... Link to comment Share on other sites More sharing options...
Guest newbe5 Posted July 23, 2011 Report Share Posted July 23, 2011 Since Eduardo is working on the wifi / wake from sleep, I tried to add Camera.apk back and debug the camera, but it wouldn't show up in the list of apps no matter where it was stored. Terminal emulator was FC-ing from /system/app, but it worked from /data/app. What should I copy from this build to the system.img from rebel1 to make video / 3D acceleration work there? I am talking about the AVP errors. Just don't push some things in to /system for now. This image is rejecting certain certificates of apps in /system. Resigning them with the debug keystore might get around this. For now, as I said, this ROM is just for test, use /data for now. Link to comment Share on other sites More sharing options...
Guest Hexxeh Posted July 23, 2011 Report Share Posted July 23, 2011 Has anyone got a Git repo with the .36 source in that compiles yet? I need an kernel image with GPT partitioning support enabled, and then I'll take another crack at getting ChromeOS running. Link to comment Share on other sites More sharing options...
Guest rebel1 Posted July 23, 2011 Report Share Posted July 23, 2011 Hi , I boot my test system with gpt partioned sdcard. It is enabled in the Kernel. My git is bootable . Https://github.com/rebel1 Link to comment Share on other sites More sharing options...
Guest ejtagle Posted July 23, 2011 Report Share Posted July 23, 2011 Just to let you know where i am right now... In fact, there was a bug on the wlan card... wlan is attached to the sdhci host #0, and not attached to the #1, as it is in the code i released so far. But, even if a correct that, i am getting this: <4>[ 606.098369] WARNING: at drivers/mmc/core/core.c:640 mmc_set_clock+0x24/0x50() <4>[ 606.105165] Modules linked in: ar6000 <4>[ 606.112141] [<c003968c>] (unwind_backtrace+0x0/0xf0) from [<c0069690>] (warn_slowpath_common+0x4c/0x64) <4>[ 606.119351] [<c0069690>] (warn_slowpath_common+0x4c/0x64) from [<c00696c0>] (warn_slowpath_null+0x18/0x1c) <4>[ 606.126668] [<c00696c0>] (warn_slowpath_null+0x18/0x1c) from [<c02cb084>] (mmc_set_clock+0x24/0x50) <4>[ 606.134104] [<c02cb084>] (mmc_set_clock+0x24/0x50) from [<c02cfd24>] (mmc_sdio_init_card+0x4c0/0x52c) <4>[ 606.141669] [<c02cfd24>] (mmc_sdio_init_card+0x4c0/0x52c) from [<c02d000c>] (mmc_attach_sdio+0x8c/0x328) <4>[ 606.149359] [<c02d000c>] (mmc_attach_sdio+0x8c/0x328) from [<c02cbb60>] (mmc_rescan+0x230/0x310) <4>[ 606.157115] [<c02cbb60>] (mmc_rescan+0x230/0x310) from [<c007b968>] (process_one_work+0x24c/0x3b8) <4>[ 606.165469] [<c007b968>] (process_one_work+0x24c/0x3b8) from [<c007beb0>] (worker_thread+0x220/0x3d8) <4>[ 606.173525] [<c007beb0>] (worker_thread+0x220/0x3d8) from [<c0080f8c>] (kthread+0x80/0x88) <4>[ 606.181693] [<c0080f8c>] (kthread+0x80/0x88) from [<c0034ba4>] (kernel_thread_exit+0x0/0x8) <4>[ 606.189935] ---[ end trace 504182fad897d53b ]--- <3>[ 606.198320] mmc0: error -123 whilst initialising SDIO card This error is caused by the WLAN SDIO card returning invalid configuration data (specifically, seems to be returning 0hz as the requested clock speed)... And then, the linux SDIO stack refuses (with reason) to set that transfer speed. We are talking about the communication speed between the WLan module and the tegra2 SoC, not the wireless speed. There must be something very silly missing here... Still under investigation. Regarding the PWM/backlight issue, i haven't checked it yet, but seems to be not a big deal to fix For those of you wanting to try the Camera, first don't forget to enable UVC (universal video camera) USB support on the linux kernel, and then, you can control the camera power (and you will notice the camera being recognized as an USB camera in dmsg), by mounting the sysfs filesystem, and then writing 1 or 0 to powerup/down the camera with a command like echo -e 1\n > (sysfsmountpoint)/devices/platform/shuttle-pm-camera/power_on to turn it on and echo -e 0\n > (sysfsmountpoint)/devices/platform/shuttle-pm-camera/power_on to turn it off. The camera is exposed as a V4L (video for linux) compliant device. Quite in fact, you can control several other modules using other entries: (sysfsmountpoint)/devices/platform/shuttle-pm-camera/power_on -> Camera power (sysfsmountpoint)/devices/platform/shuttle-pm-bt/power_on -> Bluetooth power (sysfsmountpoint)/devices/platform/shuttle-pm-gps/power_on -> GPS power (sysfsmountpoint)/devices/platform/shuttle-pm-gsm/power_on -> GSM power (sysfsmountpoint)/devices/platform/shuttle-pm-wlan/power_on -> WLan power (but there is a bug here) I have the GSM module installed, and i can confirm thatt the module is being recognized by the kernel and exposed as several serial interfases. I haven-t tried the Android image yet, but i would think the GSM module should be working... (probably requires a bit of configuration in Android) Well, so, there are missing 3 points here: Wlan ...let's hope i will be able to find the cause... (unfortunately, the chip is under a metal shield, so i can-t measure if it is being properly powered) ... Bluetooth... Untested, but i think it should be working... It is connected to one of the tegra2 usarts, so this is pretty standard Backlight issue with Earlysuspend Well, will continue investigation... Eduardo Link to comment Share on other sites More sharing options...
Guest rebel1 Posted July 23, 2011 Report Share Posted July 23, 2011 GSM is confirmed working under Android with the right configuration. I use it with your 2.6.36 Link to comment Share on other sites More sharing options...
Guest ejtagle Posted July 24, 2011 Report Share Posted July 24, 2011 GSM is confirmed working under Android with the right configuration. I use it with your 2.6.36 Well, some progress... I verified all the drivers used to control power... and found several inconsistencies... Now, when you write 0 to power_on entry in the associated sysfs , you actually power off the associated device... Previously, some devices were powering on instead! Regarding wifi, i hace managed to load the module successfully. With the files i am providing, you just load the module, then disable wifi using : echo -e 0\n > (sysfsmountpoint)/devices/platform/shuttle-pm-wlan/power_on And then reenable it: echo -e 1\n > (sysfsmountpoint)/devices/platform/shuttle-pm-wlan/power_on and it will work (you will notice that with dmesg) Alternatively, you can load the module after disabling and enabling... I dont know why this disabling and enabling is required... The SDIO stack in linux always recognizes the card properly, but seems that afterwards, the card enters a unstable state... BTW... You need to disable and then enable the card just once... Then , the module and the card will always work ... I will continue investigation on this, PLEASE, reconfigure and recompile your kernel WITHOUT the MMC EMBEDDED SDIO option ... That option, if selected, causes the WLan module not to work, as it is preventing the linux kernel from reading the device configuration required to make it work (the option is exactly for that, and assumes all SDIO cards do not support to be queried for the proper configuration, so the configuration should be harcoded in the kernel. That is NOT the case with the AR6002, that WILL supply the required configuration to the kernel, if the kernel is allowed to ask (and that will happen ONLY if this option is not checked) Regarding the problems with the Suspend/resume and backlight related, it is still under investigation, but i also suspect that the main cause is an improper configuration of the kernel. There are 2 options: CONSOLE_EARLYSUSPEND, and FB_EARLYSUSPEND. I suspect they are mutually exclusive... You should use one of them, not both... I believe that this is the problem with the non-working resume from suspend... But i will also continue investigation Regards, Eduardo The changelog is as follows: board-shuttle-power.c -> Marked ld4 as always on, because, if this power supply is turned off, the system will hang board-shuttle-gpu,c -> restoring fb_blank() calls. They should be there board-shuttle-sdhci.c -> sdhci.0 is the slot where WLAN is connected. Modify to reflect that shuttle-pm-*.c -> Refactored, lots of bugs fixed, unified criteria (power_on=0 will always turn off devices ... ts.rar Link to comment Share on other sites More sharing options...
Guest rebel1 Posted July 24, 2011 Report Share Posted July 24, 2011 (edited) Hi Eduardo , great work. 1. wlan under android solved 2. pwm issue solved. 3. battery status works under android When CONFIG_CONSOLE_EARLYSUSPEND is set, pwm backlight control doesn´t work Set instead CONFIG_FB_EARLYSUSPEND , then pwm control works as expected The git repo is up to date. Edited July 24, 2011 by rebel1 Link to comment Share on other sites More sharing options...
Guest ejtagle Posted July 24, 2011 Report Share Posted July 24, 2011 Hi Eduardo , great work. 1. wlan under android solved 2. pwm issue solved. 3. battery status works under android When CONFIG_CONSOLE_EARLYSUSPEND is set, pwm backlight control doesn´t work Set instead CONFIG_FB_EARLYSUSPEND , then pwm control works as expected The git repo is up to date. I would also say that you (and all the other people involved in the .36 HC port) have done an excellent, awesome and outstanding job ! :D ... I can barely believe that from nearly nothing, now most of HC features are working (specially hardware acceleration, WOW!!) ... :) ... And i am still amazed by the fact that the momentum reached was so high that was done in less than 2 weeks... Incredible! :) Is there anything else i could help with regarding kernel ? Accelerometer? Camera? Bluetooth? Greetings to all people here and thanks for letting me be part of this! Eduardo Link to comment Share on other sites More sharing options...
Guest a_appleby Posted July 24, 2011 Report Share Posted July 24, 2011 (edited) Hi Eduardo , great work. 1. wlan under android solved 2. pwm issue solved. 3. battery status works under android When CONFIG_CONSOLE_EARLYSUSPEND is set, pwm backlight control doesn´t work Set instead CONFIG_FB_EARLYSUSPEND , then pwm control works as expected The git repo is up to date. How do you build the WLAN driver from the kernel tree from your github repository? (nevermind, I built it) Can you share some more details on how to get various things like GPS, GSM and WLAN working? I would also say that you (and all the other people involved in the .36 HC port) have done an excellent, awesome and outstanding job ! :D ... I can barely believe that from nearly nothing, now most of HC features are working (specially hardware acceleration, WOW!!) ... :) ... And i am still amazed by the fact that the momentum reached was so high that was done in less than 2 weeks... Incredible! :) Is there anything else i could help with regarding kernel ? Accelerometer? Camera? Bluetooth? Greetings to all people here and thanks for letting me be part of this! Eduardo Your work is amazing, you did an extremely good job. I would really like to see the camera working under HC. It has only worked under 2.2 and it would be a nice touch to get it to work. I don't know how to enable the GSM, GPS, bluetooth and camera on the HC side. Thanks everyone for all their work. Edited July 24, 2011 by a_appleby Link to comment Share on other sites More sharing options...
Guest BillyBobJoe Posted July 24, 2011 Report Share Posted July 24, 2011 Your work is amazing, you did an extremely good job. Thanks everyone for all their work. I think this sums it up nicely, great job. You guys have all been fantastic I have no idea that between you things could be done this quickly. I can't wait until I can get this applied to a ROM to see what it can do. If any of you want to post up a boot.img, I'd love to see it in action. Billy... Link to comment Share on other sites More sharing options...
Guest a_appleby Posted July 25, 2011 Report Share Posted July 25, 2011 (edited) Well, some progress... I verified all the drivers used to control power... and found several inconsistencies... Now, when you write 0 to power_on entry in the associated sysfs , you actually power off the associated device... Previously, some devices were powering on instead! Regarding wifi, i hace managed to load the module successfully. With the files i am providing, you just load the module, then disable wifi using : echo -e 0\n > (sysfsmountpoint)/devices/platform/shuttle-pm-wlan/power_on And then reenable it: echo -e 1\n > (sysfsmountpoint)/devices/platform/shuttle-pm-wlan/power_on and it will work (you will notice that with dmesg) Alternatively, you can load the module after disabling and enabling... I dont know why this disabling and enabling is required... The SDIO stack in linux always recognizes the card properly, but seems that afterwards, the card enters a unstable state... BTW... You need to disable and then enable the card just once... Then , the module and the card will always work ... I will continue investigation on this, PLEASE, reconfigure and recompile your kernel WITHOUT the MMC EMBEDDED SDIO option ... That option, if selected, causes the WLan module not to work, as it is preventing the linux kernel from reading the device configuration required to make it work (the option is exactly for that, and assumes all SDIO cards do not support to be queried for the proper configuration, so the configuration should be harcoded in the kernel. That is NOT the case with the AR6002, that WILL supply the required configuration to the kernel, if the kernel is allowed to ask (and that will happen ONLY if this option is not checked) Regarding the problems with the Suspend/resume and backlight related, it is still under investigation, but i also suspect that the main cause is an improper configuration of the kernel. There are 2 options: CONSOLE_EARLYSUSPEND, and FB_EARLYSUSPEND. I suspect they are mutually exclusive... You should use one of them, not both... I believe that this is the problem with the non-working resume from suspend... But i will also continue investigation Regards, Eduardo The changelog is as follows: board-shuttle-power.c -> Marked ld4 as always on, because, if this power supply is turned off, the system will hang board-shuttle-gpu,c -> restoring fb_blank() calls. They should be there board-shuttle-sdhci.c -> sdhci.0 is the slot where WLAN is connected. Modify to reflect that shuttle-pm-*.c -> Refactored, lots of bugs fixed, unified criteria (power_on=0 will always turn off devices ... Hi Eduardo , great work. 1. wlan under android solved 2. pwm issue solved. 3. battery status works under android When CONFIG_CONSOLE_EARLYSUSPEND is set, pwm backlight control doesn´t work Set instead CONFIG_FB_EARLYSUSPEND , then pwm control works as expected The git repo is up to date. The WLAN doesn't work here. It fails to load data.patch.hw2_0.bin even if it's in the right place. Could anyone share the fixes for the hardware components? There isn't any ROM of any use right now with .36. I think this sums it up nicely, great job. You guys have all been fantastic I have no idea that between you things could be done this quickly. I can't wait until I can get this applied to a ROM to see what it can do. If any of you want to post up a boot.img, I'd love to see it in action. Billy... There aren't noticeable changes except for the fixed sleep & wake. However, I can post a boot.img later. This is mostly for debugging and development. Edited July 25, 2011 by a_appleby Link to comment Share on other sites More sharing options...
Guest Cass67 Posted July 25, 2011 Report Share Posted July 25, 2011 The WLAN doesn't work here. It fails to load data.patch.hw2_0.bin even if it's in the right place. Could not load any portion of wlan drivers until i loaded the module and powered wlan off and on again from sysfs insmod /system/lib/modules/ar6000.ko echo -e 0\n > /sys/devices/platform/shuttle-pm-wlan/power_on echo -e 1\n > /sys/devices/platform/shuttle-pm-wlan/power_on at that point it all kicked into life ... <6>[ 309.196015] mmc0: new SDIO card at address 0001 <4>[ 309.205284] ar6000_available <1>[ 309.207223] AR6000: Unsupported configuration opcode: 3 <4>[ 309.209164] Firmware mode set <1>[ 309.209745] AR6000: Unsupported configuration opcode: 3 <1>[ 309.209771] AR6000: Unsupported configuration opcode: 5 <4>[ 309.209944] BMI enabled <4>[ 309.213627] ar6000_avail: name=wlan0 hifdevice=0xc8946000, dev=0xc4fe2000 (0), ar=0xc4fe2440 <4>[ 309.215225] old options [0] old sleep [29] <4>[ 309.215499] Set hi_refclk_hz : Ref Clock=26000000 <4>[ 309.216236] AR6000: eeprom transfer by HOST <4>[ 309.216262] eeprom_ar6000_transfer: Enter <4>[ 309.216280] enable_SI <4>[ 309.372095] disable_SI <4>[ 309.372667] AR6000: BMISetAppStart <4>[ 309.372794] AR6000: enable HI_OPTION_TIMER_WAR <4>[ 309.373126] AR6000: firmware_transfer <4>[ 309.373188] firmware_transfer: /system/lib/hw/wlan/athwlan.bin.z77 size 93528 bufsize 94208 <4>[ 309.373610] firmware_transfer: BMILZData: len=8192, remaining=86016 <4>[ 309.408132] firmware_transfer: BMILZData: len=8192, remaining=77824 <4>[ 309.440855] firmware_transfer: BMILZData: len=8192, remaining=69632 <4>[ 309.473658] firmware_transfer: BMILZData: len=8192, remaining=61440 <4>[ 309.506495] firmware_transfer: BMILZData: len=8192, remaining=53248 <4>[ 309.540033] firmware_transfer: BMILZData: len=8192, remaining=45056 <4>[ 309.574197] firmware_transfer: BMILZData: len=8192, remaining=36864 <4>[ 309.608459] firmware_transfer: BMILZData: len=8192, remaining=28672 <4>[ 309.642548] firmware_transfer: BMILZData: len=8192, remaining=20480 <4>[ 309.676705] firmware_transfer: BMILZData: len=8192, remaining=12288 <4>[ 309.710923] firmware_transfer: BMILZData: len=8192, remaining=4096 <4>[ 309.745142] firmware_transfer: BMILZData: len=4096, remaining=0 <4>[ 309.761881] firmware_transfer: /system/lib/hw/wlan/data.patch.hw2_0.bin size 1328 bufsize 4096 <4>[ 309.775412] ar6000_init() Got WMI @ 0xc8ade200. <4>[ 309.931585] mac address = cc:d8:11:00:c6:0f <4>[ 309.931653] ar6000_init() WMI is ready <4>[ 309.931830] wmi_control_rx() : Unknown id 0x101e <4>[ 309.932270] AR6000 Reg Code = 0x60 # ifconfig wlan0 wlan0 Link encap:Ethernet HWaddr CC:D8:11:00:C6:0F BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:5 errors:0 dropped:0 overruns:0 frame:0 TX packets:4 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:301 (301.0 B) TX bytes:46 (46.0 B) I now have an interface ... all that remains is to configure wpa_supplicant to connect ... Then all that remains is to have Android do all this work :) Link to comment Share on other sites More sharing options...
Guest rebel1 Posted July 25, 2011 Report Share Posted July 25, 2011 (edited) Could not load any portion of wlan drivers until i loaded the module and powered wlan off and on again from sysfs insmod /system/lib/modules/ar6000.ko echo -e 0\n > /sys/devices/platform/shuttle-pm-wlan/power_on echo -e 1\n > /sys/devices/platform/shuttle-pm-wlan/power_on at that point it all kicked into life ... # ifconfig wlan0 wlan0 Link encap:Ethernet HWaddr CC:D8:11:00:C6:0F BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:5 errors:0 dropped:0 overruns:0 frame:0 TX packets:4 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:301 (301.0 B) TX bytes:46 (46.0 B) I now have an interface ... all that remains is to configure wpa_supplicant to connect ... Then all that remains is to have Android do all this work :) Here a few tips: Add this to your init.rc # Workaround : ar6000 write /sys/devices/platform/shuttle-pm-wlan/power_on 0 write /sys/devices/platform/shuttle-pm-wlan/power_on 1 write /sys/devices/platform/shuttle-pm-wlan/keep_on_in_suspend 1 <-- When you don´t want that the card is powered of when system goes to suspend Also i attached my patched libhardware_legacy.so, with this Android loads the driver an configure wpa_supplicant. The path for module and atheros blobs is with this: /system/lib/hw/wlan/ rebel1 libhardware_legacy.so.zip Edited July 25, 2011 by rebel1 Link to comment Share on other sites More sharing options...
Guest Cass67 Posted July 25, 2011 Report Share Posted July 25, 2011 Here a few tips: Add this to your init.rc # Workaround : ar6000 write /sys/devices/platform/shuttle-pm-wlan/power_on 0 write /sys/devices/platform/shuttle-pm-wlan/power_on 1 write /sys/devices/platform/shuttle-pm-wlan/keep_on_in_suspend 1 <-- When you don´t want that the card is powered of when system goes to suspend Also i attached my patched libhardware_legacy.so, with this Android loads the driver an configure wpa_supplicant. The path for module and atheros blobs is with this: /system/lib/hw/wlan/ rebel1 libhardware_legacy.so.zip Thanks for the file .. my patching of the libhw legacy fails :) Link to comment Share on other sites More sharing options...
Guest newbe5 Posted July 25, 2011 Report Share Posted July 25, 2011 Having some issues with the battery on the Vegta. Seems to work on the POV, but the Vega seems to be different: http://pastebin.com/dzphp7DN That's me shoving it through insmod, if anyone has some time to shed light on this it would be appreciated, working with rebel1 (this is his game ;)) newbe5 Link to comment Share on other sites More sharing options...
Guest ejtagle Posted July 25, 2011 Report Share Posted July 25, 2011 Having some issues with the battery on the Vegta. Seems to work on the POV, but the Vega seems to be different: http://pastebin.com/dzphp7DN That's me shoving it through insmod, if anyone has some time to shed light on this it would be appreciated, working with rebel1 (this is his game ;)) newbe5 <3>[ 4.580210] nvec nvec.0: NvEC returned an error: 0x04 <3>[ 4.580221] nvec nvec.0: NvEC not found That is the problem... NVEC is required to properly register to be able to get battery data... I may say that NVEC (Nvidia embedded controller) is a very crappy piece of hw... Error 0x04 means that the NVec is there, but does not recognize the command. Perhaps the NVEC is not using the same fw version than the NVEc in POV... look at nvec.c:1358 -- That is command that is failing ... (Get firmware version) Try to comment out lines 1365,1366 and 1367, recompile and check that the NVec driver is loading properly... I had problems with the GetFirmware version command. Perhaps previous firmware versions of NvEC didn't implement it at all... Link to comment Share on other sites More sharing options...
Guest rebel1 Posted July 25, 2011 Report Share Posted July 25, 2011 Ok, i give it a try and report back. Link to comment Share on other sites More sharing options...
Guest the_corvus Posted July 25, 2011 Report Share Posted July 25, 2011 Oh, I go out for 3 days and you get almost all working :P I will do some test tomorrow... For the next dont forgot to activate it in sys fs. For BT. Not working now, but, we need to modify init.harmony.sh in ramdisk (take bluetooth service lines from vegacomb), copy bluecoat6.psr (from system/etc in vgacomb) and bccmd (if i remember correctly from /system/bin/... you know... from vegacomb). Then you will get the BT adapter up (executing at hand at least, but doesnt work into android). For Camera. Once activated in sys, see dmesg to see if it's detected as a usb device. Then launch camera app and last time i try i get errors related with asx or asf files (firmware files). Anyway, anyone knows the exact model of our camera? Corvus. Link to comment Share on other sites More sharing options...
Guest ejtagle Posted July 25, 2011 Report Share Posted July 25, 2011 Oh, I go out for 3 days and you get almost all working :P I will do some test tomorrow... For the next dont forgot to activate it in sys fs. For BT. Not working now, but, we need to modify init.harmony.sh in ramdisk (take bluetooth service lines from vegacomb), copy bluecoat6.psr (from system/etc in vgacomb) and bccmd (if i remember correctly from /system/bin/... you know... from vegacomb). Then you will get the BT adapter up (executing at hand at least, but doesnt work into android). For Camera. Once activated in sys, see dmesg to see if it's detected as a usb device. Then launch camera app and last time i try i get errors related with asx or asf files (firmware files). Anyway, anyone knows the exact model of our camera? Corvus. Regarding the camera, it is an USB standard camera. This could differ significantly from the Acer... One of the tegra2 SoC peripherals allows the designer to connect a digital camera, and allows to drive it directly, without using USB ... and, that peripheral does also implement autofocus and video processing. But P10AN01 does not use it. Instead, they use an standard USB camera that does not require firmware at all. Once the linux kernel recognizes it, it is exported as an standard Video 4 linux. So, though it should be easy to make it work, perhaps the Acer people have written an *.so library that is trying to access the (unused) Tegra2 SoC camera interface peripheral, and trying to load firmware for it... Perhaps, we can get a *.so library that only uses standard V4L interfaces ... In such case, it should work.. Link to comment Share on other sites More sharing options...
Guest the_corvus Posted July 26, 2011 Report Share Posted July 26, 2011 That is the problem... NVEC is required to properly register to be able to get battery data... I may say that NVEC (Nvidia embedded controller) is a very crappy piece of hw... Error 0x04 means that the NVec is there, but does not recognize the command. Perhaps the NVEC is not using the same fw version than the NVEc in POV... look at nvec.c:1358 -- That is command that is failing ... (Get firmware version) Try to comment out lines 1365,1366 and 1367, recompile and check that the NVec driver is loading properly... I had problems with the GetFirmware version command. Perhaps previous firmware versions of NvEC didn't implement it at all... With this lines commented, i get the same errors... In nvec.0 all files have 0 as value. Corvus. Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now